ANGELINA, ANGELINA and Agtriadi, Herman Bedi and Djamain, Yasni (2025) PERBANDINGAN KLASIFIKASI POLA TIDUR UNTUK MEMPREDIKSI RISIKO INSOMNIA MENGGUNAKAN ALGORITMA SVM DAN XGBOOST. Diploma thesis, ITPLN.
202131152_Angelina_Revisi_Skripsi_Angelina Sudung.pdf
Restricted to Registered users only
Download (2MB)
Abstract
Klasifikasi pola tidur memiliki peran penting dalam memprediksi risiko insomnia yang dapat berdampak pada produktivitas, kesehatan fisik, serta kesejahteraan mental. Penelitian ini bertujuan membandingkan kinerja algoritma Support Vector Machine (SVM) dan Extreme Gradient Boosting (XGBoost) dalam mengklasifikasikan pola tidur untuk prediksi risiko insomnia. Data yang digunakan berasal dari Sleep Health and Lifestyle Dataset pada Kaggle, yang telah melalui tahap pembersihan, transformasi fitur, dan penyeimbangan kelas. Evaluasi model dilakukan menggunakan metrik akurasi, presisi, recall, dan F1-score. Hasil penelitian menunjukkan bahwa SVM dengan parameter C=10, gamma=0,01, dan kernel RBF memperoleh akurasi 83%, presisi 0,87, serta recall 0,77 untuk kelas insomnia. Sementara itu, XGBoost dengan konfigurasi n_estimators=50, max_depth=10, dan learning_rate=0,1 mencapai akurasi 80%, presisi 0,80, dan recall 0,81. Temuan ini mengindikasikan bahwa SVM lebih unggul dalam aspek presisi sehingga lebih sesuai untuk aplikasi klinis yang membutuhkan diagnosis akurat, sedangkan XGBoost lebih sensitif dalam mendeteksi kasus insomnia sehingga lebih tepat digunakan pada screening skala besar. Penelitian ini diharapkan dapat berkontribusi pada pengembangan sistem deteksi dini gangguan tidur, sekaligus mendukung pencapaian Sustainable Development Goals (SDGs) poin ke-3: Good Health and Well-being.
Sleep pattern classification plays an important role in predicting the risk of insomnia, which significantly affects productivity, physical health, and mental well being. This study aims to compare the performance of the Support Vector Machine (SVM) and Extreme Gradient Boosting (XGBoost) algorithms in classifying sleep patterns for insomnia risk prediction. The data used in this study were obtained from the Sleep Health and Lifestyle Dataset on Kaggle, which underwent preprocessing including data cleaning, feature transformation, and class balancing. Model evaluation was carried out using accuracy, precision, recall, and F1-score metrics. The experimental results indicate that SVM with parameters C=10, gamma=0.01, and an RBF kernel achieved an accuracy of 83%, a precision of 0.87, and a recall of 0.77 for the Insomnia class. Meanwhile, XGBoost with configurations n_estimators=50, max_depth=10, and learning_rate=0.1 obtained an accuracy of 80%, a precision of 0.80, and a recall of 0.81. These findings suggest that SVM outperforms in terms of precision, making it more suitable for clinical applications that require high diagnostic accuracy, while XGBoost demonstrates higher sensitivity in detecting insomnia cases, making it more appropriate for large-scale screening. This study contributes to the development of early detection systems for sleep disorders and supports the achievement of the Sustainable Development Goals (SDGs), particularly Goal 3: Good Health and Well-being.
| Item Type: | Thesis (Diploma) |
|---|---|
| Uncontrolled Keywords: | Pola Tidur, Insomnia, Support Vector Machine, XGBoost, Machine Learning, Sleep Pattern, Insomnia, Support Vector Machine, XGBoost, Machine Learning |
| Subjects: | Skripsi Bidang Keilmuan > Teknik Informatika |
| Divisions: | Fakultas Telematika Energi > S1 Teknik Informatika |
| Depositing User: | Sudarman |
| Date Deposited: | 15 Oct 2025 03:56 |
| Last Modified: | 15 Oct 2025 03:56 |
| URI: | https://repository.itpln.ac.id/id/eprint/2336 |
